Two basic objectives for any device intended for automatic measurement of coagulation times are, first, that an established endpoint in the coagulation process be precisely and immediately detected so that significant reproducible results will be obtained and, second, that there be no cross contamination or carry-over from one sample to the next. While some devices known in the past have satisfied the first of these objectives, and others the second, such devices have generally failed to achieve both. Thus, electrical coagulation detectors which measure changes in conductivity (resistance) as an indicator of clot development, and optical detectors which respond to changes in reflected or transmitted light, may present relatively few carry-over or contamination problems but, unfortunately, they generally lack precision, reliability, and reproducibility in endpoint detection. On the other hand, mechanical or electro-mechanical detectors, which usually include immersible elements and which respond more directly to changes in viscosity, are more precise in endpoint detection but clearly present substantial carry-over problems.
An early method for measuring the coagulation time of freshly-drawn whole blood was that of H. Vierordt [1878, Arch. f (physiologische) Heilkunde Vol. 19] who disclosed drawing a natural filament (a meticulously cleaned white horse hair) through a glass tube containing raw blood and determining the coagulation endpoint by observing when the filament became reddish (because of red cells clinging thereto) or the coagulum (fibrin) attached itself to the moving filament. In 1910 K. Kottmann and A. Lindsky [Zeitschzift fur Klinische Medizin 69:431-434 (1910)] modified Vierordt's method by incorporating thermoregulation, and in 1921 R. Schmidt [Medizinisch Klinik 17. Jahrgang Nr. 16, 439-60 (1921)] altered the thermoregulation aspect by using a U-tube instead of the straight tubes used by Vierordt and by Kottmann and Lindsky.
While Vierordt's approach may still be suitable for so-called global assays in which the coagulation time of freshly drawn whole blood is determined, and in which the coagulation time is measured in minutes, it is not suitable for the more specific coagulation tests, such as prothrombin time tests and activated partial thromboplastin time tests, where results are measured in seconds, and even in fractions of seconds, following the addition of a coagulating agent to a plasma sample. Despite considerable activity in this field, prior efforts have failed to produce a precise and reliable automatic method for measuring coagulation times, particularly for the specific assays so widely used in clinical laboratories.
The following references, and the citations therein, further indicate the state of development of the prior art: U.S. Pat. Nos. 3,766,774, 3,658,480, 3,605,010, 3,267,364, 3,268,804, 3,704,099, 3,267,363, 3,020,748, 3,077,106, 3,038,327, 3,518,057, 3,560,162, 3,525,254, 3,458,287, and 3,492,096.